Forensic Comparison of Soil Samples Using Nondestructive Elemental Analysis

Area of Science:

• Forensic Science
• Geochemistry
• Analytical Chemistry

Background:

• Soil analysis is crucial for linking evidence in forensic investigations.
• Traditional methods may risk sample integrity (e.g., microfossil destruction).
• Need for rapid, reliable soil comparison techniques.

Purpose of the Study:

• To develop and validate an adapted ED-XRF method for forensic soil analysis.
• To optimize data processing and statistical comparison for soil samples.
• To demonstrate the practical application of the method in real-world scenarios.

Main Methods:

• Adapted Energy-Dispersive X-ray Fluorescence (ED-XRF) analysis, utilizing sieving to preserve microfossils.
• Elemental composition data generated for 20 elements.
• Evaluation of data processing techniques and statistical distances (e.g., Canberra distance) using log-LR cost (C_llr).

Main Results:

• Identified optimal data processing: Canberra distance, relative data, and square root values.
• Constructed a discriminative model demonstrating high comparative accuracy.
• Successful application in twelve test cases, confirming method applicability and spatial resolution.

Conclusions:

• The adapted ED-XRF technique, combined with optimized data processing and a discriminative model, provides a rapid and effective tool for forensic soil comparison.
• The method preserves sample integrity, allowing for detailed microfossil analysis alongside elemental profiling.
• This integrated approach enhances the utility of soil as forensic evidence.